Reference guide for token counting and prefix caching across LLM providers (OpenAI, Anthropic, Gemini). Use when debugging token counts or optimizing prefix caching.
Install
npx skills add letta-ai/letta --skill llm-provider-usage-statisticsnpx skills add letta-ai/letta --skill llm-provider-usage-statistics -a claude-code -gnpx skills add letta-ai/letta --skill llm-provider-usage-statistics -a codex -ggemini skills install https://github.com/letta-ai/letta.git --path .skills/llm-provider-usage-statisticsgh skill install letta-ai/letta llm-provider-usage-statisticsgit clone --depth 1 https://github.com/letta-ai/letta /tmp/letta-ai__letta
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/letta-ai__letta/.skills/llm-provider-usage-statistics ~/.claude/skills/llm-provider-usage-statistics # Claude Code
cp -R /tmp/letta-ai__letta/.skills/llm-provider-usage-statistics ~/.agents/skills/llm-provider-usage-statistics # Codex, Cursor, Gemini CLI, Copilot, OpenCode